Using conventional manual methods for inspection of building images is a time-consuming, costly process, and the results often contain inconsistent standards of examination. Because it is crucial to understand building surface conditions and make maintenance decisions at an appropriate time, the construction industry has been developing an automated defect and damage classification method. The objective of this study is to obtain low-cost and high-quality images from digital cameras for building damage detection experiments. The researchers conducted sample training and testing through artificial intelligence technologies and later analyzed the testing results to evaluate the performance of supervised machine learning methods for concrete efflorescence detection. The support vector machine (SVM) enables clearly distinguishing differences between normal concrete and concrete with efflorescence and the results classification indicated the most satisfactory assessment performance. Analysis indicated that the efflorescence scalar was 56.7% and the efflorescence vector was 53.1% in the study. The quantity of digitized surface damage could indicate the extent of building degradation and provide an initial reference for estimating damage scope and severity.